Xiamen
Joined May 2015
- Tweets506
- Following255
- Followers2.4K
- Likes417
In December 2021, a year before ChatGPT, I asked whether computers could run the simulations, analyze the results, and write the papers by themselves.
My full lecture on where that now stands:
youtu.be/zUgz0r3nldU
we might or might not write a paper about this very efficient fine-tuning route to accurate ML interatomic potentials (often works even on your CPU machine, without GPU!). First, we wanted to release it ASAP to make it available to the community.
MLatom 3.25: ultra-efficient fine-tuning of its universal machine learning potentials (AIQM, OMNI-P, ANI, ...) on your own data.
On ethanol: 35 geometries, 1.37 → 0.67 kcal/mol against its target, 10-30 s per run.
pip install --upgrade mlatom
#compchem #mlchem #opensource
Science Done on a Machine by a Machine: #AIAgents in #CompChem
Perspective by me, Hassan Nawaz, and Arif Ullah: arxiv.org/abs/2608.18508 (if we missed your agents, let us know, and we can update our later versions)
Recently, I have decided to change MLatom's license to Apache 2.0 to make it even more permissive than it was (MIT with the citation clause). In addition, installation has been simplified too, just a single command pulling all the most needed dependencies.
MLatom is now Apache 2.0 — permissive, commercial-friendly, patent grant included, and the old citation clause is gone.
One line on Python 3.9–3.11:
pip install -U mlatom
Source: github.com/dralgroup/mlatom
Docs: aitomistic.com/mlatom
#compchem #mlchem #opensource
just a small example of how all future #compchem calculations will look like. Not just the future, but the present is already agentic.
Find a transition state — just by chatting.
One prompt → an NEB path → the TS → its imaginary mode → IRC → the reaction energy diagram. Submitted, monitored & analyzed for you. No commercial QM; powered by AIQM3.
Free: aitomistic.xyz
#compchem #AIagents #AI4Science
Insane how far #compchem has jumped in a year. A year ago, we had Aitomia — standard simulations (geom opt, freq, …). Today: Protomia (aitomistic.com/protomia) does all that and drafts & edits manuscripts, writes code, and everything else you'd expect from a top agentic system.
Our framework, Aitomia, is now published in JCTC!
"Aitomia: An Agentic Framework for AI-Driven Atomistic and Quantum Chemical Simulations" doi.org/10.1021/acs.jctc.6c0…
Want the live, ready-to-use system? Try Protomia on the Aitomistic Hub: aitomistic.com/protomia
Here is a reminder about Faraday Discussion poster abstract deadline. I will be giving an invited talk at Faraday’s discussions - might be a good chance to meet some of you in person!
invt.io/1txby0pyxoi
OMNI-P2x is now out in Nat. Commun.! The first universal neural network potential for excited electronic states of small molecules. Approaching TD-DFT accuracy at lower cost. Lots of room to improve; fine-tuning helps. @mikolajmartyka
doi.org/10.1038/s41467-026-7…
Paper detailing FSSH implementation in MLatom and some cool functionality with hybrid QM/ML models, flexibility, versatile stop function, and interesting benchmark comparisons is now out in @JCIM_JCTC : doi.org/10.1021/acs.jctc.5c0….
#compchem #mlchem
See also cool new tutorials by Jakub Martinka, who has done the FSSH implementations: aitomistic.com/mlatom/tutori…
AIQM3 @JCIM_JCTC : more elements, good performance for drug-design-related applications (better than tested universal MLIPs), and applicability to both charged and radical species, while preserving good performance for transition states.
doi.org/10.1021/acs.jctc.5c0…
#compchem #mlchem
You can try out AIQM3 online on the Aitomistic Hub at aitomistic.xyz .
Paper: doi.org/10.1021/acs.jctc.5c0…
Tutorial: mlatom.com/docs/tutorial_aiq…
Our novel approach of directly predicting nuclear positions for various molecules as a function of time, rather than doing stepwise propagations as in molecule dynamics, is finally published in @JCIM_JCTC :
doi.org/10.1021/acs.jctc.5c0…
#compchem #mlchem #moleculardynamics
Our recent @ChemicalScience article ‘AIQM2: organic reaction simulations beyond DFT’ ( pubs.rsc.org/doi/D5SC02802G ) has been listed in the journal's 2025 most popular machine learning and automation articles collection.
#ChemSciMostPopular #compchem #mlchem #aichem
Our publication on enabling trajectory surface hopping with electronic structure methods without analytical gradients with #ML is out in @ChemicalScience!
Read more about it:
doi.org/10.1039/D5SC09557C
Detailed tutorials, etc., are coming soon. Stay tuned!
#compchem #mlchem
A great piece by @robinson_julia in
@ChemistryWorld on how #AIagents will democratize #compchem. Soon, manual QC inputs will feel like building pyramids. Students already start by chatting with Aitomia at aitomistic.xyz. Gen-2 coming soon.
tinyurl.com/53ybzs9h
Pavlo Dral retweeted
Have you ever wanted to use cutting-edge #compchem methods, such as QD-NEVPT2, to propagate NAMD but were stopped by the lack of available energy gradients? If so, check out our new preprint by @mikolajmartyka , J. Jankowska, H. Lischka, and
@PavloDral : doi.org/10.26434/chemrxiv-20…
It is my great pleasure to be a subject chair of #RSCPoster 2026! Looking forward to learning about your great science through your digital posters!
Very humbled to see our research with @JakubMartinka , Lina , @mikolajmartyka , Yi-Fan, Jiri, and @MarioBarbatti among the most-read recent articles in @JPhysChem Lett.
Paper: doi.org/10.1021/acs.jpclett.…
My personal account of the study’s background:
dr-dral.com/jpcl-a-descripto…
I am looking forward to participating in Faraday Discussion's Molecular excited states theory and experiment, 14-16 September 2026, Cambridge, UK, rsc.li/excitedstates-fd2026
Deadline for Oral abstract submissions is 15 December 2025.
#compchem #aichem #mlchem
